ISTAT - Istituto Nazionale di Statistica
Euro-SDMX Metadata Structure Definition (ESMS_MSD 3.0 ESTAT)
Statistics on the production of manufactured goods
2014 - A0
1. Contact
1.1 Contact organisation
Istat - Italian National Statistical Institute
1.2 Contact organisation unit
SSI/C - Structural Business and Institution Statistics Division / Statistics on production and intermediate consumption unit
1.3 Contact name
1.4 Contact person function
1.5 Contact mail address

Via Tuscolana, 1788 00173 Rome Italy

1.6 Contact email address
1.7 Contact phone number
1.8 Contact fax number
2. Metadata update
2.1 Metadata last certified
03/03/2016
2.2 Metadata last posted
03/03/2016
2.3 Metadata last update
03/03/2016
3. Statistical presentation
3.1 Data description

PRODCOM is an annual survey of the value and volume of products produced by industrial enterprises in Italy and sold during the reference year. It is carried out in compliance with EU Council Regulation on the Establishment of a Community Survey of Industrial Production (No.3924/91).

The PRODCOM survey provides information on PRODuction of the European COMmunity and is compiled from data received from italian manufacturers. The survey sample is approximately 41.000 entreprises, covering 234 industries, derived from sub-sectors of the total manufacturing sector, and about 3900 products. Data available include national main indicators and their definition (value and volume of product sales, total production in volume).

3.2 Classification system

In the PRODCOM classification products are grouped by the NACE Rev.2, 4-digits industrial sectors, in which they are primarily produced (i.e. characteristic products). The value and volume data shown for a product, therefore, includes not only the production by enterprises classified to the NACE sector where it is primarily produced but also its subsidiary production by enterprises classified to the other NACE Rev.2 sectors.

Classification of Products:

The PRODCOM list is divided into “Classes” corresponding to the classes of NACE rev 2. Within the individual classes, the headings of the list are set out in ascending order of their PRODCOM codes. PRODCOM uses an eight-digit product code of the form XX.XX.XX.YY. The first four digits of the code correspond to the 4-digit classes of NACE Rev. 2. The first six digits are the Community Classification of Products by Activity (CPA) codes. The CPA provides a detailed listing of the characteristic products for each 4-digit NACE Rev. 2 economic activity. The last two digits provide a more detailed breakdown of the CPA classes into PRODCOM product headings. There is a direct link between the PRODCOM classification and the EU foreign trade Combined Nomenclature (CN). Each PRODCOM 8-digit heading covers one or more CN 8-digit headings.

At national level:

To view the national classification, please see ATECO2007 (http://www.istat.it/it/archivio/17888)

3.3 Coverage - sector

The survey covers all industrial enterprises with 3 or more employees, which are wholly or primarily engaged in industrial production and industrial services in the Mining, Quarrying, and Manufacturing industries.

Under Prodcom, the NACE Rev2 classification is having covered the sections B and C, comprising sections 07-33, but excepting 09 and 19.

In order to fulfil the regulation provisions (to reach 90% representativeness, Art.3), a probabilistic sample based survey strategy has been designed. It involves a sample of small industrial enterprises (from 3 to 19 employees) and the totality of units that belong to medium and large companies (20 employees and over).

The sampling method is based on an equiprobabilistic selection in each stratum, obtained by the combination of two variables: class of employees and class of economic activity. Variables that are used to evaluate the sample size in each stratum are: turnover, added-value and number of employees. The size is obtained by applying the Bethel method.

3.4 Statistical concepts and definitions

Either Sold Production (the value and volume of the product sold by the enterprise) or Total Production (the volume of all production of the product, including both the proportion that is sold and the proportion that is retained by the enterprise for adding to stocks, using in further processing etc.). Where Total Production is required, Sold Production by value and volume must be reported as well.

In Italy it is investigated also:

    - Production further processed or incorporated into another product of the same local unit during the reference period (volume);

    - Production further processed or incorporated into another product in another local unit of the same company during the reference period (volume);

    - Production realised in the local unit on contract processing (as contractor) (volume);

    - Fee paid by the principal to the contractor for production carried out on contract processing;

    - Sold production that has been processed on contractual-basis to other companies (production sold by principals) (volume)

3.5 Statistical unit

The reporting unit is Enterprise. The observation unit is Local Unit(s).

The results are compiled at enterprise level before being published. We add up the amounts related to each enterprise.

3.6 Statistical population

The survey population is made up of all industrial enterprises in NACE 07-33 with more than 20 persons and a sample of small industrial enterprises (from 3 to 19 employees).

3.7 Reference area

Italy - The whole national territory.

3.8 Coverage - Time

The period observed is one calendar year.

Italian industrial products statistics date back to 1996.

Electronically, data from 1996 onwards are available at http://ec.europa.eu/eurostat/web/prodcom/data/excel-files-nace-rev.1.1.

For 1996 to 2007, the files contain the original data based on NACE Rev.1.1. For 2008 onwards, the files contain data that has been converted where possible from the data based on NACE Rev.2.

3.9 Base period

Not applicable.

4. Unit of measure

Italy is using the same measurement units as the ones from Eurostat list (http://ec.europa.eu/eurostat/cache/metadata/Annexes/prom_esms_an3.doc)

5. Reference Period

Year 2014.

6. Institutional Mandate
6.1 Institutional Mandate - legal acts and other agreements

European Union law:

- Regulation No.3924/1991: EU Regulation (EEC) No.3924/91 of 19 December 1991 on the establishment of a Community survey of industrial production

National law:

The process is included in the National Statistical Program (PSN) that is annually approved by Decree of the President of Republic (DPR) on the basis of the Law no.125 of 30 October 2013 (http://www.gazzettaufficiale.it/eli/id/2015/11/05/15A08158/sg ). The statistical survey is mandatory to be filled-in by the private units. More info about national law regarding statistics here: http://www.istat.it/it/istituto-nazionale-di-statistica/organizzazione/normativa.

6.2 Institutional Mandate - data sharing

Not applicable

7. Confidentiality
7.1 Confidentiality - policy

Several  legal acts are present in Italy to guarantee the confidentiality of data requested to enterprises for statistical purposes.

Italian full policy on data confidentiality is available at http://www.istat.it/it/istituto-nazionale-di-statistica/organizzazione/normativa/normativa-nazionale.

The Statistical law of Italy stipulates that following data must be confidentialised:

    1. individual statistical data;

    2. statistical data which can be matched in a way that enables the identification of a specific statistical unit;

    3. statistical information which aggregates data about less than 3 statistical units or about a population in which the relative share of the value of a surveyed parameter of a single unit exceeds 85% of the total value of such parameter for all units in the population.

7.2 Confidentiality - data treatment

The Statistics Italy's confidentiality policy for Prodcom data stipulates that a certain Prodcom code will be flagged where there are less than 3 reporting companies (for each published indicator).

8. Release policy
8.1 Release calendar

The provisional data for 2014 are available at http://dati.istat.it (published in July 2015).

The revised data will be available starting 23 Jun 2016.

8.2 Release calendar access
8.3 Release policy - user access

The information is available on line on http://www.istat.it/it/archivio/12026

9. Frequency of dissemination

Data are disseminated at national level on yearly basis.
Also, data are provided to Eurostat on yearly basis (provisional Prodcom data released on June 30, along with the final figures of the previous year).

10. Accessibility and clarity
10.1 Dissemination format - News release

None.

10.2 Dissemination format - Publications
10.3 Dissemination format - online database
10.4 Dissemination format - microdata access

As part of microdata release, Istat offers a research data centre: ADELE Laboratory (Laboratory for Elementary Data Analysis), a “secure” site accessible by researchers and academics enabling them to conduct their own statistical analyses on microdata from the Institute's surveys in compliance with legislation concerning the confidentiality of personal data (http://www.istat.it/en/products/microdata-files)

10.5 Dissemination format - other

Tables: Statistica annuale della produzione industriale. Anno 2005 http://www.istat.it/it/archivio/13524

Tables: Statistica annuale della produzione industriale. Anno 2006 http://www.istat.it/it/archivio/13432

Tables: Statistica annuale della produzione industriale. Anno 2007 http://www.istat.it/it/archivio/9815

Tables: Statistica annuale della produzione industriale. Anno 2008 http://www.istat.it/it/archivio/12602

Tables: Statistica annuale della produzione industriale. Anno 2009 http://www.istat.it/it/archivio/43647

Tables: Statistica annuale della produzione industriale. Anno 2010 http://www.istat.it/it/archivio/73150

Tables: Statistica annuale della produzione industriale. Anno 2011 http://www.istat.it/it/archivio/102585

10.6 Documentation on methodology

On webpage of online database there is a short description of methodology. (http://dati.istat.it/?lang=en, and choose in left-upper side the theme "Industry and Construction", then "Production" and "Production volume").

10.7 Quality management - documentation

The I.Stat Information system on quality of statistical production processes SIQual (http://siqual.istat.it/SIQual/lang.do?language=UK) contains information about the execution of Istat statistical production processes and about activities developed to guarantee the quality of the produced statistical information. There are no ESMS files available at national level since for the moment these are not widespread.

11. Quality management
11.1 Quality assurance

Since 90s Istat adopted a systematic approach to ensure quality in both statistical information and service to the community. For this purpose, the Institute has defined a quality policy providing itself with appropriate tools as well as management changes to carry it out.

Istat quality policy is coherent with the European framework developed by Eurostat, taking up its main principles and definitions. In 2005, the European Statistics Code of Practice (revised in 2011) established the principles to follow in order to ensure and strengthen both accountability and governance of the European Statistical System and the National Statistical Systems within it.

Essential points of Istat quality policy are:

  - Process quality: consisting in the production of accurate statistical information efficiently and effectively;

  - Product quality: consisting in the dissemination of high-quality timely statistical data which are relevant for the users, also the potential ones;

  - Documentation: consisting in the storage and availability of information necessary not only for a proper use of data but also to ensure transparency in all the production activities of statistical data;

  - Respect for respondents: consisting in the reduction of response burden and in the respect of respondent's privacy;

  - Strengthening of statistical literacy: consisting in promoting a proper use of statistical information in policy-making to better support decisions and policies;

  - Users' orientation: consisting in making statistical information easily accessible and understandable and in satisfying user needs as much as possible.

For details: http://www.istat.it/en/about-istat/quality

11.2 Quality management - assessment

The Information System provides a number of arithmetic and logical controls, both mandatory and optional between the items in each form and between the forms in the whole annual report. There are a number of arithmetic and logical controls inside the PRODCOM form. The System calculates both the totals and the unit price. The PRODCOM questionnaire can not be completed successfully before all errors inside the questionnaire, detected by the checking program, to be corrected.

12. Relevance
12.1 Relevance - User Needs

The main users of PRODCOM data can be presented in different groups, namely:

a) EXTERNAL/ INTERNATIONAL (OUTSIDE NSI): Eurostat;

b) EXTERNAL/ NATIONAL (OUTSIDE NSI), like:

     -  Institutions at national level - governamental (Ministry of Economy; Ministry of Agriculture and Food, ENEA the Italian National Agency for New Technologies, Energy and Sustainable Economic Development);

     -  Institutions at national level - non-governamental (Trade associations; Researchers and students, Companies);

c) INTERNAL (INSIDE NSI): Environmental statistics, National accounts,  STS,  Agriculture statistics

12.2 Relevance - User Satisfaction

No user satisfaction survey has been done for Prodcom.

12.3 Completeness

In Italy are inquired all codes from of the Procom list.
Quality indicator: 100%.

13. Accuracy
13.1 Accuracy - overall

For 2014, the response rate was of 46% of all enterprises surveyed.

The main causes of non-response are: misclassification of economic activity (activities outside of sections B and C, Nace Rev.2), idle activity, transformation (merger or separation), stop of activity, impossibility to contact the company.

After checking if the company has stopped or suspended its activity during the reference year, the non-respondenses were estimated using the previous years data, by applying the annual variation of the main STS indicators. In case of partial responses, that concerns basically missing volume data, estimates for missing volumes are done by using the median unit price from the other respondent companies.

13.2 Sampling error

A probabilistic sample has been designed. It involves a sample of small industrial enterprises (from 3 to 19 employees) and the totality of the local units that belong to medium and large companies (20 employees and over).

The sampling method is based on an equiprobabilistic selection in each stratum, obtained by the combination of two variables: class of employees and class of economic activity. The variables used to evaluate the sample size in each stratum are: turnover, added-value and number of employees. The size is obtained by applying the Bethel method.

In order to ensure adequate coverage, productive local units belonging to companies having main activities outside the industrial sectors covered by Prodcom are included in the survey as well. The required coverage levels for Prodcom survey is achieved by means of a grossing up technique. The estimation technique currently adopted makes use of a calibration method based on the auxiliary variables number of enterprises and number of employees belonging to industrial companies included in the target population. The sampling CV (coefficients of variation) of each statistic (turnover, added-value and number of employees) is estimated by the ratio between square root of sampling variance estimate and parameter estimate in each estimation domain.

CV = 0.05 for each statistic (turnover, added-value and number of employees)

13.3 Non-sampling error

Coverage error:
- The frame for Local units is Italian Statistical Business Register (ASIA), managed by Istat (National Statistical Institute of Italy)
ESS standard over coverage rate: edition 31/12/2014 = 3,69% (alpha=1.00)

The quality actions to control non-response are:
- Survey presentation letter signed by Istat President
- Guarantees on statistical confidentiality
ESS standard unit nonresponse rate: edition 31/12/2014 = 53,95% (alpha=1.00)

Editing and imputation procedures are used to correct the non-response and measurement errors.

Data are validated to the following actions:
- Coherence control with data from other surveys or sources
- Coherence control with previous data of the same survey
- Control of influential unit data, strongly contributing to computation of aggregates (e.g. large enterprises)

14. Timeliness and punctuality
14.1 Timeliness

Data are disseminated at national level on a yearly basis, and are provided to Eurostat on a yearly basis also.

Timeliness indicators are:

Edition Timeliness (days) Timeliness of Provisional Data (days) Timeliness of First Release for Data Under Revision Policy (days)

Timeliness for Eurostat  (days)

Timeliness for Istat NA, other sectors or Italian Statistical System (days)
31/12/2014   201   181  
14.2 Punctuality

In Italy, the Prodcom provisional data are usually published without delay in relation to the scheduled date. Information is available at http://www.istat.it/it/archivio/12026.

15. Coherence and comparability
15.1 Comparability - geographical

Not applicable.

15.2 Comparability - over time

Length of comparable time series:

VariableLengthNote
Total Production from 01/1996 to today  
Sold production (volume) from 01/1996 to today  
Sold Production (value) from 01/1996 to today  

 

An extensive revision of NACE in 2007 has resulted in changes to PRODCOM estimates for the 2008 survey onwards.

Every 5th year, there is a revision of the HS (Harmonized System) which will lead also to more significant changes to CN and Prodcom. The latest major revision was in 2012.

Revision in NACE will also affect Prodcom. Prodcom data from 1995­ to 2007, based on NACE Rev.1.1, have been backcasted to NACE Rev.2 by Eurostat.

15.3 Coherence - cross domain

Not applicable.

15.4 Coherence - internal

Not applicable.

16. Cost and Burden

Average company's time to fill in the survey: 2.8 hrs.
Actions to reduce burden: pre-filled questionnaire and single sign (access with a unique userid and password), using enterprise portal.
Just for the sample we use a coordinated selection of Jales technique (a sampling technique).

17. Data revision
17.1 Data revision - policy

Data released at the beginning of July are preliminary, and revisions are disseminated after 12 months.

After 12 months the data can be considered as final, unless huge mistakes in time series are detected in data submitted to Eurostat.

17.2 Data revision - practice

When submitting the Prodcom data for a new year to Eurostat, Statistics Italy will submit also the revised versions of previous years, when is relevant. Typically, the previous years have significant revisions to mandate a revised submission.

As difference, the mean is (-8,293,000) and weighted mean with the last sold value is (2,327).

18. Statistical processing
18.1 Source data

The survey sample is selected from Business Register ASIA. PRODCOM estimates cover Italy’s registered businesses for over 90% (we made census for enterprises with 20 or more employees, which totals over 30.000 units).

For the enterprises with less than 20 employees we made a sample of about 10.000 units (while the population in Italy is over 175.000). VAT and turnover is classified to NACE divisions 07-33. Businesses are selected by employment size and industry classification, using a stratified random sample. Additional constraints are built into the sample design so to decrese the form filling response burden on smaller businesses. The number to be selected has been calculated to minimise the burden of sampling, without compromising the calculation of final estimates.

The sampling method is based on an equiprobabilistic selection in each stratum obtained by the combination of two variables: class of employees and class of economic activity. The variables used to evaluate the sample size in each stratum are: turnover and number of employees. The determination of numerosity sample and its allocation between the stratum occurred requiring that the expected error in the estimated total (error expressed in terms of coefficient of variation) of the variables considered were less than 5 percent in each domain of study represented by the class of economic activity. The allocation of the sample in the layers was achieved by adopting a multivariate allocation algorithm proposed by Bethel, 7 criterion, which is a generalization of the method of Neymann for only one variable, and that allows to minimize the sample size so as to satisfy the constraints precision of the estimates for each of the variables considered of interest, relative to the domain estimation.

18.2 Frequency of data collection

On a yearly basis.

18.3 Data collection

Data collection:
Self-administered data collection by e-questionnaire (Computer Assisted Web Interviewing-CAWI)
The web questionnaire and instructions for companies are available at https://indata.istat.it/prodcom/

The quality actions to control unit nonresponse are:
- Survey presentation letter signed by Istat President
- Guarantees on statistical confidentiality
- Coercive measure: the answer to the statistical process is mandatory with penalties for enterprises with over 100 employees.

18.4 Data validation

Data are validated to the following actions:
- Coherence control with data from other surveys or sources
- Coherence control with previous data of the same survey
- Control of influential unit data, strongly contributing to computation of aggregates (e.g.large enterprises)

18.5 Data compilation

Steps of data compilation in Prodcom statistics of Italy:

  1. Error detection based on consistency edits
  2. Manual error correction forcing wrong values on the base of deterministic rules as reported in a manual or document
  3. Manual error correction forcing wrong values on the base of previous correct data of the same survey
  4. Manual error correction based on reporting unit follow-up
  5. Manual coding assisted by paper lists (manuals, listings) or computer screens
  6. Data entry by Istat section in charge of the survey
  7. Outsourcing data entry
  8. Splitting questionnaire according to various record types
  9. Deterministic error and outlier detection and imputation based on deterministic rules (IF-THEN)
  10. Error and outlier detection and imputation by clerical reviewers based on experience
  11. Error and outlier detection and imputation by clerical reviewers based on data from previous editions of the same survey
  12. Error and outlier detection and imputation by clerical reviewers based on questionnaire review
  13. Error and outlier detection and imputation by clerical reviewers based on respondent follow-up
  14. Deterministic error and outlier detection and imputation by clerical reviewers based on experience
  15. Deterministic error and outlier detection and imputation by clerical reviewers based on data from previous editions of the same survey
  16. Deterministic error and outlier detection and imputation by clerical reviewers based on data from other surveys or sources
  17. Deterministic error and outlier detection and imputation by clerical reviewers based on questionnaire review
  18. Deterministic error and outlier detection and imputation by clerical reviewers based on respondent follow-up
  19. Deterministic error and outlier detection and imputation by clerical reviewers based on the available auxiliary variables
  20. Estimate calculation
18.6 Adjustment

Not applicable for Prodcom data.

19. Comment

No comment.